Bolt Debugging Prompt for Designers
You get a short list of likely causes ranked by probability, plus the exact checks to run that confirm or rule out each one. This prompt is tuned for designers working in Bolt: it carries your role context, and the steps below cover what changes in Bolt. Check that research summaries reflect what participants actually said, not patterns the AI expected, by going back to the raw notes for any key finding.
Help me debug a problem. I want to find the cause, not just patch the symptom. ## Symptom [what you see, how often, and since when] ## Environment [language, framework, versions, OS, local or production] ## Relevant code and logs [paste the code involved and any log output or stack trace] ## What changed recently [deploys, dependency updates, config changes, or "nothing I know of"] ## What I have already checked [and what each check showed] ## Instructions - List the likely causes, most probable first, and say why each fits the evidence. - For each cause, give one specific check (a log line to add, a command to run, a value to print) that confirms or rules it out. - Do not propose a fix until we have narrowed it down, unless the cause is obvious. ## Output format A numbered list: cause, why it fits, how to check it. Then one line on which check to run first. About my work - Product and user: [e.g. scheduling app for clinic front-desk staff] - Screen or flow: [e.g. rescheduling an appointment on mobile] - Design system and constraints: [e.g. our button and form components, WCAG AA] Context - App and screens involved: [framework, page or component names] - Leave unchanged: [features, routes or styling that must not change]
How to use this prompt
- 1
Open Bolt
Paste the prompt into Bolt's chat to start or update a project.
- 2
Give it context
Say which framework and features the app has, and which files or screens the change affects.
- 3
Review and iterate
Check the live preview after each step and describe what is wrong in plain terms, one problem at a time.
Working in Bolt
- •Bolt builds and runs the whole app in the browser, so a vague prompt can change more than you intended. Scope each request.
- •Ask it to keep existing features working and to list the files it changed.
- •For errors, paste the exact message from the preview or console instead of describing it.
Worked example
What you give it
Symptom: add_tag returns tags from earlier calls. Calling add_tag("a") then add_tag("b") gives ["a", "b"] on the second call instead of ["b"].
Environment: Python 3.11, local script.
def add_tag(tag, tags=[]):
tags.append(tag)
return tags
What changed: nothing, it has always done this.What a good answer looks like (shortened)
1. Mutable default argument (very likely). The list [] is created once, when the function is defined, and every call without a tags argument reuses that same list.
Check: print(id(tags)) inside the function. If the id is the same on every call, this is the cause.
2. A caller is passing the same list in each time (less likely, since the calls you showed pass only one argument).
Check: search for calls to add_tag that pass a second argument.
Run check 1 first. If confirmed, the fix is:
def add_tag(tag, tags=None):
if tags is None:
tags = []
tags.append(tag)
return tagsTips for better results
- •Describe the screen state by state (empty, loading, error, success) when asking for UX copy or specs. It forces the edge cases that handoffs usually miss.
- •Write "What I have already checked" with results, not just actions. "Restarted the server, error still happens on the first request" rules out more than "restarted the server."
- •For errors, paste the exact message from the preview or console instead of describing it.
- •Avoid this: Leaving out what changed recently, which is often the fastest clue to an issue that started suddenly.
Check before you use it: Check that research summaries reflect what participants actually said, not patterns the AI expected, by going back to the raw notes for any key finding.
Get the free prompt pack
Our best coding prompts in one Markdown file, delivered by email. The templates here stay free either way — the pack just saves you the copying.
Variations
Intermittent bug
Add this line to Symptom: It only happens [e.g. 1 in 20 requests, under load, after midnight UTC]. Focus on timing, concurrency and caching causes.
Production only
Add this line to Instructions: I cannot attach a debugger. Suggest only checks that work through logs, metrics or config.
Add logging
Change Output format to: The exact log statements to add, where to put them, and what each output would tell us.
Save it as a Retext snippet
Retext is a free text expander for Mac. Save this prompt once, then type ;debug in any app (ChatGPT, Claude, your editor, a browser tab) and the full prompt is pasted in.
- Click Add to Retext. The prompt installs as a snippet with the trigger ;debug.
- Change the trigger in Retext if you prefer a different one.
- Type the trigger wherever you want the prompt, then fill in the brackets.
Frequently asked questions
Can AI help with accessibility reviews?
It can check copy, heading order, labels and color contrast values you give it, and list common issues to test. You still need to test with a screen reader and keyboard, since it cannot see your actual interface.
What is the difference between debugging and bug fixing prompts?
A debugging prompt is for when you do not know the cause yet, so it asks for hypotheses and checks. A bug fix prompt is for when you can reproduce the problem and want a patch.
How do I keep Bolt from rewriting the whole app?
Name the screen or file to change and add "Leave everything else as it is." Smaller requests give smaller, easier to check diffs.
How do I reuse this prompt without copying it every time?
Click Add to Retext to save it as a snippet. Then type ;debug in any Mac app, including your AI tool, and the full prompt is pasted in.
Related Templates
Get the free prompt pack
Our best coding prompts in one Markdown file, delivered by email. The templates here stay free either way — the pack just saves you the copying.